Batch Compute
- Product Introduction
- Getting Started
- Console User Guide
- User Guide
- Command Line Interface
- Practical Tutorial
- API Documentation
- Compute Environment APIs
- Making API Requests
- Configuration Viewing-related APIs
- Task Template-related APIs
- Job-related APIs
- SDK Documentation
Quick Start
Last updated: 2024-01-13 11:19:29
Overview
This document describes the usage methods and computing capability of Batch.
Prerequisites
Complete preparations based on the instructions in Preparation, and learn how to configure the general part of the custom information.
Viewing the Demo
Note:
Modify the general part of the custom information in
1_SimpleStart.py based on the instructions in Preparation.Open
1_SimpleStart.py in an editor.# custom (Change to your info)imageId = "img-m4q71qnf"Application = {"DeliveryForm": "LOCAL","Command": " python -c \\"fib=lambda n:1 if n<=2 else fib(n-1)+fib(n-2); print(fib(20))\\" "}StdoutRedirectPath = "your cos path"StderrRedirectPath = "your cos path"
All information except
Application in the general part is described in Preparation. Set the parameters of Application based on the following table.Configuration Item | Description |
DeliveryForm | Three methods are available for delivering an application: software packaging, container image, and direct running within a CVM instance. LOCAL indicates direct running within a CVM instance. |
Command: | The command for starting a job. In this example, a Python script is executed. Starting from 1, the script adds up the first 20 numbers of the Fibonacci sequence and outputs the result to StdOutput. |
Submitting a Job
Run the following command to run the Python script.
The demo encapsulates the job submission process by using Python scripts and the Batch command line tool.
python 1_SimpleStart.py
The returned result is as follows, indicating that the job is successfully submitted:
{"RequestId": "393292f4-5583-48ad-a9f5-f673138ea637","JobId": "job-o0xxxxxq7"}
Viewing State
Run the following command to view the execution status through DescribeJob:
$ tccli batch DescribeJob --version 2017-03-12 --JobId job-xxx
Note:
The returned result is as follows (some information is omitted):
{"EndTime": "2019-10-08T04:06:58Z","JobState": "SUCCEED","TaskInstanceMetrics": {...},"Zone": "ap-guangzhou-2","TaskMetrics": {...},"JobName": "TestJob","Priority": 1,"RequestId": "7a5f4c94-1357-486c-9c48-8286ba01b5b2","TaskSet": [...],"StateReason": null,"JobId": "job-o0xxxxxq7","DependenceSet": [],"CreateTime": "2019-10-08T04:05:54Z"}
The common values of JobState in the returned result are as follows:
STARTING: Launching
RUNNING: Running
SUCCEED: Running successfully
FAILED: Failed to run
Viewing the Result
1. Log in to the COS console. In the left sidebar, click Bucket List.
2. Click the ID of the desired bucket, go to the object list page, and view the log files that store the execution results. See the figure below:
If the job succeeds, view the standard output in stdout.job-xxx.xxxx.0.log. The content is as follows:
6765
If the job fails, view the standard error in stderr.job-xxx.xxxx.0.log. The content may be as follows:
/bin/sh: -c: line 0: syntax error near unexpected token `('/bin/sh: -c: line 0: ` python -c \\"fib=lambda n:1 if n<=2 else fib(n-1)+fib(n-2); print(fib(20))\\" '
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No